Interpreting Kerala’s COVID-19 Numbers
Through a series of data visualisations, the authors attempt to describe Kerala's COVID-19 trajectory, the state’s policy response and how to assess its pandemic performance. Using demographic, economic, social (particularly public health) data, the authors provide context on the vulnerability of the state to outbreaks, show how to use comparative indicators, and account for district-level variation and reporting errors. Finally, the authors examine what has changed in the consequent waves of the pandemic and discuss the enduring strengths of the “Kerala model.”
On 30 January 2020, Kerala became the first Indian state to report a case of the novel coronavirus. However, four days before this, the Government of Kerala (GoK) had already released COVID-19 guidelines that established case definitions, screening protocol, hospital preparedness and triage. The implementation of these measures delayed transmission from imported cases for over 40 days until 9 March, when Kerala saw its first cluster outbreak in the Pathanamthitta.
Figure 1: Case Trajectory and Major GoK Policy Measures (30 January to 8 May 2020)
Within a week, the GoK initiated the “Break the Chain” campaign to communicate information on hygiene and physical distancing. Then, a ₹20,000 crore relief package was announced, days before a national relief package. On 22 March (Day 51), the GoK imposed a statewide lockdown starting the following day, that is, a day before the national lockdown. On 28 March, Kerala reported its first COVID-19 death and issued advisories on the creation of first-line treatment centres and COVID-19 care centres. By the end of April, Kerala made wearing face masks mandatory in public. On 6 May (Day 98), Kerala’s doubling rate of infections was high at every 346 days. What followed were two days of zero reported cases, marking the end of the first wave of the pandemic. The next day (Day 101), the first repatriation flight returned with 359 expats from Abu Dhabi and Dubai, initiating the second wave of the virus through imported transmissions.
Figure 2: Trends in Cases and Institutional Measures
Over the following hundred days, cumulative cases grew exponentially from 503 to nearly 43,000. Testing, isolations and hospitalisation (of symptomatic patients) grew on an overall basis but fell on a per-detected case basis, indicating that Kerala had begun to hit capacity constraints. For instance, it can be seen from Table 1, while the state was conducting 71 tests (Routine-PCR) per detected case during the first wave of the virus, this number fell to 12 in the next 100 days and has remained at three ever since. Similarly, home isolation and hospitalisation (of symptomatic patients) also fell on a per active/confirmed case basis. This is indicative of the under-testing and relaxation of containment measures for the returning non-resident Keralites (NRKs) (Chathukulam and Tharamangalam 2021).
By June 2020, officials began observing positive cases that were not linked to those with a travel history. This trend extended to fatalities as well, where those without virus symptoms were found to be COVID-19 positive posthumously. By 1 September, 200 such clusters were identified in various areas ranging from coastal villages to the hilly areas, from markets to old-age homes and factories. On 24 November, 300 days into the pandemic, Kerala had 64,000 active cases and five persons in home isolation per active case. The death rate had been maintained below 0.4%. As of February 2021, Kerala accounts for the largest proportion of active COVID-19 cases in India. How does one interpret these numbers? Does this indicate a failure of Kerala’s globally lauded response to the pandemic?
In this article, we contextualise Kerala’s deteriorating pandemic performance with other socio-economic data of the state, to argue why Kerala is uniquely susceptible to a public health crisis and, therefore, why the easing of certain restrictions led to such a large upsurge in cases.
Table 1: Inter-wave Comparison on Select Indicators
|
|
100th Day (8 May) |
200th Day (16 Aug) |
300th Day (24 Nov) |
Current (Jan 23) |
|
Cumulative cases |
503 |
43,912 |
5,71,872 |
8,84,242 |
|
Active cases (as on) |
16 |
15,349 |
64,412 |
72,048 |
|
Cumulative deceased |
3 |
153 |
2,095 |
3,587 |
|
Case fatality rate |
0.60% |
0.35% |
0.37% |
0.41% |
|
Cumulative tests (routine) |
35,886 |
5,11,471 |
18,41,744 |
26,57,226 |
|
Tests per detected case |
71 |
12 |
3 |
3 |
|
Total home isolations (as on) |
19,810 |
1,48,793 |
2,98,902 |
1,99,889 |
|
Number isolated per active case |
1,238 |
10 |
5 |
3 |
|
Cumulative hospitalisations (of symptomatic patients) |
|
59,852 |
15,850 |
11,935 |
|
Hospitalisation per detected case |
13 |
1 |
0.03 |
0.01 |
Source: Compiled from Kerala Dashboard and the Directorate of Health Services
Characteristics Unique to Kerala
First, in many ways, Kerala is more demographically and socially susceptible to pandemics.
Aging population with high NCD morbidity: Kerala has a fertility rate of 1.8 (SRS Statistical Report 2017), which is below replacement rate of 2.1 and one of the slowest decadal population growth rates in the country (4.9% versus 17.7% nationally). As a result, it has one of the oldest populations in the country.
Figure 3: Percentage of Population over 60 in India
Kerala also has a high level of morbidity arising from non-communicable diseases, accounting for 52% of deaths, as opposed to 42% in India; one in five adults has diabetes and one in three adults has hypertension, almost double the national rate, and incidence of cancer and strokes is much higher than in the rest of India (AMCHSS 2016-17). Given that individuals over 60 and those with comorbidities tend to be more vulnerable to COVID-19, the state is at a relatively higher risk of case fatalities.
Large diaspora and international air traffic: There are over 2 million emigrants who have left Kerala to work abroad. Almost 90% of them are in the Gulf nations (Rajan 2020), though there are significant Keralite populations in North America and Europe. In 2018–19, Kerala had over 10 million people pass through the state’s four international airports, the third highest rate in the country (Airport Authority of India 2020), indicating Kerala’s exposure to the risk of imported transmission. Although the first lockdown eliminated this risk to a large extent, as repatriation flights were initiated and later regular flights resumed, testing and contained measures imposed on arriving air passengers were inadequate.
Figure 4: International Air Traffic of Different States
Urbanised and densely populated state: Kerala is a highly urbanised state, with 47.7% of its population living in urban centres versus the national average of 33%. Kerala has one of the highest population densities in the nation (860 people per square kilometre), almost thrice the national average, thereby increasing the risk of local and community transmission once there is imported transmission. While urban density need not automatically result in higher transmission rates, as we learnt from the East Asian experience (Abraham and Tandel 2020), it does in Indian context, due to a number of factors related to urban planning such as housing (indoor crowding).
Figure 5: Population Density by State
Second, when making relative assessments of states, one must use comparable indicators.
Population adjusted figures: In the following analysis, we compare Kerala to other major states over the first 300 days of the pandemic (until 24 January 2021). For the first 200 days, the cases lagged far behind other states. Even while cases sharply spiked in the latter quarter of 2020, deaths still significantly lagged behind other states, on a per million basis.
Figure 6: Cases per Million in Comparison States
Figure 7: Deaths per Million in Comparison States
Case adjusted figures: Test positivity rate (percentage of positive results out of total tests) and case fatality rate (percentage of fatalities out of total positive cases) are good comparable indicators to assess relative state performance. Kerala performed better than other major states on both accounts during the first and second hundred days of the pandemic. While the test positivity rate has increased over time, it remains low compared to other states. Case fatality rate has remained constantly low, compared to other states.
Figure 8: Cumulative Positivity Rate in Comparison States
Figure 9: Cumulative Case Fatality Rate in Comparison States
Third, given the level of administrative autonomy at the district level, there is a significant variation in non-COVID-19 health indicators, COVID-19 transmission and consequently policy response among Kerala’s districts.
For instance, after a few initial outbreaks in Kasargod, the district implemented triple lockdown. Wayanad, one of the least populous districts of Kerala, has been isolating the most number of persons (adjusted for population), in part because the doctor-turned-IAS district collector Adeela Abdulla prioritised quarantining efforts and enforcement delayed geofencing (Cris 2020).
Figure 10: District-level Variation in Selected Health Indicators
Figure 11: District Level Cases and Institutional Effort
Fourth, when analysing Kerala COVID-19 figures, one must pay attention to the units of and errors in reporting. For example, the GoK dashboard reports the total number of persons in home isolation on any given day. Because the period of quarantine is not fixed, it is impossible to calculate the number of persons who began their isolation period on any given day. Further, what is reported as “Hosp Today,” to the best of the authors’ understanding, is the number of symptomatic persons under institutional isolation and not hospitalisation from severe symptoms. Finally, it has also been previously found that fatality numbers are prone to mis-categorisation and under-reporting in the state (Kurien 2020), something that must be accounted for when reporting death numbers.
A final consideration when assessing policy response is the trade-off between the economic and public health consequences of imposing or lifting restrictions, which cannot be understated in the Kerala context. Tourism, a sector hard hit by the pandemic, accounts for 10% of Kerala’s gross domestic product and 23.5% of its employment (Nair and Dhanuraj 2018). Moreover, the state receives over ₹85,000 crore in household remittances annually forming 19.3% of the state domestic product (Rajan and Zachariah 2020).
Figure 12: Remittances per Capita
The loss of livelihoods of NRKs has an immediate implication on the state’s economy, where at least 16% of households depend on remittances for income. As a result, the state that was the first to go into lockdown also became one of the first to lift restrictions. Weddings and funerals, first restricted and later allowed, became sources of COVID-19 outbreaks, as did the celebration of festivals such as Onam. Lockdown fatigue also began to set in, not just among citizens but also among front-line workers and other government functionaries, who had worked intensive shifts for months on end.
Yet, marked by its incredible coordination efforts and decentralised decision-making, the Kerala governmental response to the pandemic continues to be unparalleled and noteworthy. The government has been able to rapidly mobilise its functionaries into different working groups and evolve a consultative decision-making process that also included stakeholders (such as experts, political opposition, volunteers) external to the government.
Figure 13: Various GoK and External COVID-19 Coordinating Bodies and their Composition
As the pandemic continues to unfold, the need of the hour is not just state policy and state action, but also eliciting public trust and citizen engagement. Therein lies one of the greatest strengths of the “Kerala model” made evident by the state’s Sannadhasena Platform for volunteer workers. As on 24 February, the platform has registered over 3,77,308 volunteers across the various districts, demonstrating the ability of the state to organise a participatory response through community action.
Conclusions
When interpreting Kerala’s COVID-19 numbers, it is important to contextualise the data to the demographics and other characteristics of the state, use comparative indicators, account for district-level variation and reporting errors. Although the current situation seems in stark contrast to the early months of the pandemic, it does not take away the state’s early success in rapidly mobilising resources, creating institutional mechanisms for coordination, continuously evolving protocols and engaging with citizens.
Must Read
